Predictors of long-term mortality in Multicenter Automatic Defibrillator Implantation Trial II (MADIT II) patients with implantable cardioverter-defibrillators.

BACKGROUND: Data on long-term follow-up and factors influencing mortality in implantable cardioverter-defibrillator (ICD) recipients are limited.

OBJECTIVE: The aim of this study was to evaluate mortality during long-term follow-up and the predictive value of several risk markers in the Multicenter Automatic Defibrillator Implantation Trial II (MADIT II) patients with implanted cardioverter-defibrillators (ICDs).

METHODS: The study involved U.S. patients from the MADIT II trial randomized to and receiving ICD treatment. Data regarding long-term mortality were retrieved from the National Death Registry. Several clinical, biochemical, and electrocardiogram variables were tested in a multivariate Cox model for predicting long-term mortality, and a score identifying high-, medium-, and lower risk patients was developed.

RESULTS: The study population consisted of 655 patients, mean age 64 +/- 10 years. During a follow-up of up to 9 years, averaging 63 months, 294 deaths occurred. The 6-year cumulative probability of death was 40%, with evidence of a constant risk of about 8.5% per year among survivors. Median survival was estimated at 8 years. Multivariate analysis identified age >65 years, New York Heart Association class 3-4, diabetes, non-sinus rhythm, and increased levels of blood urea nitrogen as independent risk predictors of mortality. Patients with three or more of these risk factors were characterized by a 6-year mortality rate of 68%, compared with 43% in those with one to two risk factors and 19% in patients with no risk factors.

CONCLUSION: A combination of a few readily available clinical variables indicating advanced disease and comorbid conditions identifies ICD patients at high risk of mortality during long-term follow-up.